How Czechia Flattened the Curve (Maybe, Hopefully)


Our current prime minister has been in the past often criticized as akin to Donald Trump re: conflict of interests and use of state resources to enrich himself and his family. And rightly so in my opinion, I cannot stand the man personally and politically.

However, when SARS-CoV-2 hit the Czech Republic, he, unlike Donald Trump, has done the right thing. In response to the pandemic, he has left decisions on the policy to actual epidemiology experts from the very beginning. Thus when CZ had mere 116 cases, 12 days after the first three on March 1., he declared a state of national emergency and just two days later virtually everything was put on hold except the absolute bare minimum (grocery stores, delivery services, apothecaries and some more). It was criticized by the opposition (our equivalent of US conservatives) as needless panic-making and fearmongering and the measures as needlessly draconian and a PR for himself and his party. Especially the order of mandatory face masks (home-made and improvised masks are allowed) was met with scorn.

On March 18. I have taken the data of confirmed cases so far, plotted them on a graph and calculated the best-fit exponential curve. It was at a daily increase of 39%, an effective doubling every two-three days, approsimately the same trajectory it has had all over Europe. This growth meant we should have over 140.000 cases today, but we, luckily, do not. We have less than 5.000. Howso?

Look at this graph:

The red curve is the actual cumulative cases as reported every day at midnight. The blue curve is the exponential best fit that I have calculated on March 18. And then there is the orange curve, which is also an exponential best-fit but only for the last week from March 28. to April 3. You can see that the two best-fit lines intersect on March 21.-22.

That is, in my opinion, the day when the enacted measures started to have a visible effect – eight to ten days after they were enacted. I do not know whether I am doing the right thing here mathematically – I have dabbled in statistics at work, but not in epidemiology – but it does seem right to me.

The new rate of growth is still exponential, but instead of 38% daily it is 8% daily. And although the difference between multiplying the cases daily by 1,08 instead of 1,39 does not intuitively look like much, it means the doubling of the cases is prolonged from mere 2-3 days to 10-11 days. Still not enough for an illness that can take up to 6 weeks to heal and kills 1% of infected people, but a very noticeable drop.

And AFAIK that drop is not due to insufficient testing. Testing has grown proportionally, although still not as much as it perhaps should have. But the ratio between positive/negative tests is getting lower, and that indicates that the drop in overall cases is real.

Now there is certainly much more to it than this oversimplified graph. For example, Germany took longer to enact strict active measures, relatively speaking. That is, CZ government enacted nation-wide strict measures when we had just several hundred people ill, whilst the German government did leave many decisions to individual states and instead of strict orders tried to control the situation with recommendations only at first. This has led to a bit of inconsistent reaction and different measures being enacted (and ignored by people) in different states. It worked, but not as much as was desired. Strong nation-wide measures started being implemented only when there were several thousand people ill already- at about the same time as in CZ. And at about the same weekend the curve began to break in Germany as well.

It was similar in Italy too, there the curve began to break at around March 15. (only estimated, I did not calculate the fit curves for Italy, I am doing this in OpenOffice and that is not the best program for this kind of work), about two weeks after the most-hit municipalities were put on lock-down.

Another quick analysis that can be done just by looking at the numbers – In Italy, it took 22 days for the cases to grow from about 100 to 20.000. In Germany, it took 24 days, in Spain 18 days, in UK and France 25 days and in the USA 20 days. The Czech Republic is now 24 days from its 100th case and we are nowhere near 20.000.

So even these amateurish and quick&dirty analyses show that quick reaction, regardless of what the nay-sayers say, is essential in avoiding the worst in case of an epidemic. The enacted measures work as intended. I only hope that our government and our people do not relax too soon.

Stay safe, stay at home whenever possible, and fingers crossed for you and your loved ones.

Comments

  1. lumipuna says

    Congrats!

    I immediately noted my own derision at the mask thing, which we then discussed to some extent. Now, it seems global expert opinion is gradually shifting, while mask shortage remains a problem.

    It’s probably still early and otherwise difficult to say which particular elements in this scheme have been practically significant. In Finland we seem to have settled on similarly slow growth with much less stringent measures -- but maybe that’s partly a result of cultural factors and low population density.

  2. says

    Latvia reached 100 confirmed cases on 20th March. Now it is at 533 cases.

    I guess you probably already know about this site https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6 but I will mention it anyway. At the bottom right corner you can find graphs of confirmed cases and daily increase. Comparing those between countries is interesting. Some (like the USA) have ridiculously steep increases. Others (like Czechia or Latvia) have much slower increase of new cases.

    Obviously, by now it should be no surprise that countries with at least semi competent politicians who quickly enacted various measures that slowed down the spread of the virus are doing much better than those that tried to pretend that everything is great in order to not disturb the stock market.

    And here https://aatishb.com/covidtrends/?location=Canada&location=China&location=Czechia&location=Estonia&location=France&location=Germany&location=Iran&location=Italy&location=Japan&location=Latvia&location=Lithuania&location=Russia&location=South+Korea&location=Spain&location=US&location=United+Kingdom is another interesting graph.

  3. rq says

    What worries me is, since the increase in cases is quite slow (good!) and people (also medical professionals) are getting bored (bad!), starting to see media commentary about how these measures are too drastic and we’re ruining the economy, and you cannot compare us to Italy or covid to the flu blah blah blah… it’s quite distressing because there is not enough information about how “NOTHING IS HAPPENING” is the goal.

  4. says

    One thing is that I think that “confirmed cases” are a very bad measurement to go by. Given the nature of this disease which doesn’t seem to show any ore distinctive symptoms in half of those infected, mild symptoms in another 25%, but severe to fatal outcomes in the remaining 25% with some yet unexplained and unpredictable factors involved, the number of confirmed cases hinges a lot on the number of tests a country is running and also the time it takes to return the results.
    For example, Germany has steadily increased its testing capacities and also reduced the time it takes to get results. two weeks ago the waiting time was one week. Now it’s about 2-3 days, which means that a backlog of cases is now showing up in the statistics. It also means we’re catching many mild cases, which is one reason why Germany has such a low death rate.
    Does anybody believe that the number of confirmed cases in Germany can be compared to those in the USA? I don’t think so. I think that the only numbers you can currently compare is hospital admittance and death. While they, to reflect infection rates from1-3 weeks ago, they are easily counted.
    Nevertheless, it looks like Czechia is doing a great job.
    But yeah, the talk about “returning to normal” for the sake of the economy” and even virologists who should know better talking about “controlled infection” of the population are frightening.
    For one thing, while people like to use war metaphors, this isn’t one. Nothing has been destroyed. No house has collapsed, no factory has burned down. Everything is still here, we can just “switch it on” again.
    The other thing: “Controlled infection?” FFS. What if I’m the unlucky one who dies? What if my kid is the rare case of a child dying? How are you supposed to protect the “risk groups” when they’re 30% of the population?
    And I also want to punch all those who talk about “herd immunity”. Herd immunity is NEVER the result of an infection running its course through the population. Herd immunity happens when a vaccinated population protects those that cannot be vaccinated, hopefully eradicating the disease over time. What you get with a disease running its course is infection waves as you have with the measles. Sure, for a time the number of those without antibodies is so low that the disease cannot take hold, but then it just comes back. Apart from the fact that they don’t even know yet whether having had it offers any actual protection and for how long. In the worst case you have something like herpes that will be with you forever.

  5. says

    Giliell @#4

    I think that the only numbers you can currently compare is hospital admittance and death.

    If we want to be pedantic, these numbers can also be misleading.

    How many of the infected people die depends on whether hospitals can cope with the number of sick people in their region.

    How many people get hospitalized depends on how much free space is available in hospitals. In some regions hospitals admit also patients with medium severity symptoms. In other regions, there are too few hospital beds even for very sick patients.

    For example, right now in Latvia there are 2 dead people, 542 confirmed cases, 21 453 people have been tested. 44 people are in hospitals, but among those only 5 people have severe symptoms and only 3 require ventilators (for now the number of infected people is still small enough that hospitals have free space also for patients with medium severity symptoms).

    Of course, given how some infected people have no symptoms whatsoever, it’s obvious that the real number of infected people in Latvia is larger than that of confirmed cases. Still, I doubt that it can be much larger. Here testing has been sufficient for weeks by now, results are usually available in two days.

  6. says

    If we want to be pedantic, these numbers can also be misleading.

    How many of the infected people die depends on whether hospitals can cope with the number of sick people in their region.

    Sure, but I think they correlate with actual numbers much better than “confirmed cases”

  7. says

    @Giliell, Death rates in Germany and Czechia are quite similar actually -- 1,57-1,58% of confirmed cases. The difference in testing is about 25% -- Germany has circa 11.000 tests per million, Czechia has circa 8.000 tests per million.

    What if I’m the unlucky one who dies? What if my kid is the rare case of a child dying?…What you get with a disease running its course is infection waves as you have with the measles. Sure, for a time the number of those without antibodies is so low that the disease cannot take hold, but then it just comes back.

    As far as I am able to understand epidemiologists, this is unfortunately inevitable. A controlled epidemic is, at this moment, a lesser of two evils. The other evil is uncontrolled one and where that leads we see in Italy and soon we will get a much better example in USA.

    Unfortunately, at this time, containment is not possible anymore. All we can do is to stall until a vaccine and/or a curativum is developed.

  8. says

    For one thing, you cannot have a “controlled epidemic”. That’s the very nature of an epidemic. How would that work? You put all know cases into quarantine and then you draw lots and the loser is infected? We’re currently slowing the epidemic. Ideal would be to contain it. And stalling for a vaccine and/or effective treatment is not a controlled epidemic where we’re all just waiting to catch it.
    I think whoever suggests that we need to have a “controlled” epidemic and that “60% of the population getting it is inevitable” (which would still be up to a million deaths in Germany. Right now we see the death rates you get without medical rationing) should put themselves and their families first. Let’s see how much they believe in “it’s inevitable” when you ask them to bring their spouses and kids to the Corona ward without any PPE.

  9. says

    Giliell, what do you think should be done except slowing the spread of the disease as much as possible to flatten the curve as much as possible -- which is, AFAIK what epidemiologists mean when they say “controlled epidemic”? There is a difference between “60% population gets it in a few months” and “60% population gets it spread over several years”. There is also a difference between “60% population gets it, mainly those at low risk of complications” and “60% population gets it, completely at random”.

    Because it really is not possible to indefinitely stop all activities all around the worlds and lock everyone at home without social contact until the virus dies out -- which could take months (nevermind that the virus can apparently now spring back to animals, so there will probably emerge non-human reservoirs, as is the case with many other diseases).

    And I am not talking about economy here, I am talking about things that definitively must be done, and on schedule, otherwise, the damage is irreversible and non-trivial. Like planting crops and harvesting them. Delivering goods to people. Healthcare (other than Covid-19 related). Education -- you cannot say that schools will be completely closed for a year or two and that it is no biggie. It is a biggie if children miss education and you know better than most that it cannot be fully supplemented at home. And the list can go on.

    Even if Czechia, for example, manages to stop the spread completely, it is just a matter of time after that before it is imported again from some other region.

    Not to mention that after a month or two of nothing much happening people will get restless. It is inevitable. There are already people breaking the quarantines and rules, and the longer the lockdown is, the more they will do so. People are like that.

    The virus is here and it will stay here. To develop a vaccine, if it is possible at all, will take a year at best. I do not like it any more than you do since about 90% of my family including me falls into the “at-risk” category. Right now, Covid-19 is among the top possibilities for my parents and my deaths and I really think that it cannot be avoided, only delayed. I would do you a favor and die since my life is not worth much anyway, but my parents still need me.

  10. says

    There is a difference between “60% population gets it in a few months” and “60% population gets it spread over several years”. There is also a difference between “60% population gets it, mainly those at low risk of complications” and “60% population gets it, completely at random”.

    So how do you make that second thing happen? Right now, with the most restrictive measures in place, very vulnerable people still get it. This virus is cutting through care facilities like a hot knife through butter. How do you protect 40% of the population while the remaining 60% somehow slowly gets through Covid 19?

  11. voyager says

    It’s hard to get federal numbers for Canada, but Ontario provincial numbers seem to show that we’re nowhere near flattening the curve.
    All of the people in my circle are staying home and taking precautions, but on Saturday night, one of my neighbours down the street had a hot tub party. Jack and I passed the house on our walk and I was gobsmacked. There were so many cars that people were parking on the lawn.

  12. says

    @Giliell, I am not above contradicting experts when I think I have strong evidence on my side -- I did call out on my czech blog our prominent sexologist when he denied the existence of bisexual people based on nothing more than that it is his opinion, for example -- but in this case, I am well aware of my limitations.

    I wasn’t happy when our head epidemiologist said that given the development in the world, complete eradication through confinement and quarantines is no longer feasible and that we as a society must prepare for recurring bouts of Covid-19 in coming years until a significant percentage of the population gets it. He has changed his mind on this in the last week in light of the evidence. I am not qualified to give advice on policy. Our head epidemiologist and his team of experts are currently working on what they consider to be the best way forward and I personally will do what I normally do in such circumstances -- follow the advice of the experts.

    I won’t play an armchair general to someone who not only has access to much better evidence to inform his decisions but who is also several times over better qualified to interpret said evidence correctly.

Leave a Reply