COVID-19, the situation here in Belgium : 17 April, 2020, Friday

E :  (our English version of today's report in French)


These data originate in the Belgian Government's Public Health System


Sources of data : (in French at these links)

De la SPF Santé publique

Sciensano (the better choice).


Their key numbers for today:

Chiffres clés - 17 avril

Cases, deaths, hospital and ICU admits and discharges. Deaths in hospitals and Rest Homes. In the latter 92% have a diagosis of "possible COVID-19" due to lack of test result. Excuse me: lack of a test.


Their complete report (10 pages on Belgium).


Their weekly report : 

Bulletin hebdomadaire de Sciensano, 9 avril


What they call their dashboard

Le "dashboard" de Sciansano 


More useful info on Belgium from Wikipedia. Quite complete and in English with appropriate judgements.


Our add-ons :

Our little map of COVID-19 in Belgium by Region.


On this map, click on any of the four icons to reveal more details: the confirmed cases, and the percentage of cases in each of Belgium's 3 Regions. Plus a fourth for Region Unknowns and foreigners.


Regional Distribution results today 

Flanders to the west, Brussels in the middle, Wallonia in the east.


Distribution Régionale - 17 avril


New cases announced each day

This is essentially, positive tests, as with oher countries.

Two days ago,  information was provided about deaths in nursing homes. There, most who died did not have a test for SARS-CoV-2. Nevertheless, these are diagnosed as "probable COVID-19." So they are folded into the total cases, without a positive test. 

One can also appreciate a less than smooth curve, since clumps of data that had not been reported daily as things evolved, are suddenly dumped in when received.


Nouveau Cas 17 avril

Above are Total Confirmed Cases.


A smoother curve, because each day is cumulated on the preceding result. (The total of all the cases, each day).

Total des Cas Confirmés - 17 avril 2020


Growth Factor Analysis

The above data are compared each day with the previous day's level of total cases. The difference between these paired days calculates a "New Cases" estimate. The new cases on any day are divided by the new cases the previous day to give a growth factor. If new cases doubled, that would be 2.0. The goal is to get below 1.0 where viral transmission will stop, and to stay below 1.0. 


Things seemed to be gradually diminishing since confinement measures were put in place March 10 through March 13 in Belgium. Currently this lockdown is proposed until May 3.


Two days ago appeared a scary rise in growth. It most probably resulted from a clump of cases from other days, being added all on 1 day.  Garbage in. Garbage out. That's the numbers business.


Today and yesterday again seemed more consistent with previous trends.


Taux de Croissance - 17 avril 2020


The growth factor includes a combined variable for [transmission · infectivity]. (E · p).

The more contacts an infected person has, the higher the value of E.

The more transmissable the virus (Ro), the higher the probability of infection with each contact.

We'll leave it at that. (E · p) shoud get below 0 if infection is no longer happening. It is the target for confinement, social distancing, hand washing, Cloroxing surfaces and door handles, masks, etc). To reduce (E · p).


It too had a bit of difficulty swallowing data two days ago. Today is just above zero and more in keeping with recent trends.

Negative values are good, though not intuitive.


Taux de Croissance (E · p) - 17 avril 2020


The mean value for this Growth Factor today, (1.293) with its standard deviation and calculated 90% Confidence Limits looks like this below. Again, as long as it's above 1.0, exponential growth is still a possibility for upcoming days. With 90% certainty, it might be as low as 1.118 or as high as 1.468. Both are greater than 1.0, so the risk of exponential growth still exists.


Taux de Croisasnce - Lim



Are things getting better?

Is there a demonstrable response to social distancing, hygiene, etc. ?

If so, the value of (E · p) now, should be lower than at the beginning of all this.

We compared the 19 last days, with the previous 15 using statistical methods shown below, and that I'll pass on to assure your greater comfort.


Analysis of means (N


The answer is that there is no difference that statisticians would call "significant."


Now why would that be? Because there is too much variability in the sampled results. This suggests as well that there is too much variability in the carrying out and enforcing of lockdown measures. But the trend towards improvement is in fact there. All that's missing is enough passage of time (and continued efforts by individuals) to get a result that pleases even a statistician.


Watch out, here it comes agaon ...


Taux de Croisasnce - Lim



We are beginning to understand a bit more clearly, the reality of the situation, and the choices that have been made ...





Our Death Report looks like this. It shows an overall mortality figure of 14.29% .

Often reported at WHO as between 1 and 4% for this viral illness worldwide.

Some will call this the CFR or Case Fatality Ratio (deaths / cases).


On 16 March in Belgium, this value for COVID-19 cases was 0.46%. Clearly quite an increase.

Deaths in Wallonia have been increasing as well, and especially now in Liège, the provincial capital. 

Dumping a whole pile of unreported deaths in Flanders into the report on one day, helped to beat back the apparent rise in deaths in Wallonia.


Décès Résumés - 17 avril 2020

WIthout getting too far into this, it suggests that less testing has been done in Wallonia than Flanders. Also that by number of inhabitants, Brussels has had the most deaths, followed by Wallonia then Flanders.




Our following Summary takes data from Sciensano (Gov't Public Health), and tries to fill in some gaps.

Sample gaps: where a person died is not always known. The person's gender is not always know.


It does suggest an increase in male deaths over female deaths, but given missing data (summarized in box on left below) anything could be true. Worldwide, male deaths seem to predominate.


The main goal was to identify regional differences in gender and rate of testing of those who died.


Ranges of age, and many other variables (Public Health seems currently focused on hospital admissions and discharges), are included in Public Health's report, linked to above, and not duplicated here.


Not knowing a decedent's gender (30% of deaths), or where that person physically died (49 cases) may seem surprising, but real. Makes one think that no one looked underneath the sheets.


Mortality Summary od Sciensano\\\'s data - April 17, 2020


The report, (without the arrows) is linked here.




The forecast for this day.

This process is based on the Growth Rate presented above, and a fixed number of resources (like hospital beds). It guesses at the percentage of the population that is infected, how severely ill, hospitalized, in ICU, and things like that.


It looks like this :


Taux de Croissance prévisions - 17 avril 2020


When one compares it with actual reports from Public Health, it seems to overstate pretty badly. 712.35% of ICU beds are not filled. Couldn't be anyway.


So what gives?


These forecasts taken in light of reported data, teach that if hospitals are not used, they are not overwhelmed.


If you don't test for SARS-CoV-19 in Rest Homes (Rest camps as I call them), when they die in their beds in the camps, they don't die in intensive care.


Before crying out: "Shame !"

I will ask you to recall where things were a few days ago. Come back to a photo of Italy (maybe on TV, I don't watch TV). See doctors in tears in Intensive Care in Italy and now in the United States. See patients on mattresses in the hallways, dying. Listen to the discussions on how to select the patients who will be mechanically ventilated, from among all those who should be ventilated but will not be.


So, before you quickly conclude that the above numbers must just be wrong, and move on to the next available pastime, ... maybe the numbers are absolutely correct.


But remember, these are projections. And decisions taken to change the way a system works can change projections. It's always more difficult to detect and judge non-decisions taken, than decisions that lead to visible actions.


So our Intensive Care Units in Belgium remain clean, fairly calm, and not submerged in chaos and death.


People in Belgium are witnessing (or living through unawares), a period of difficult decisions. But the above figures are also witnesses.


It is always easier, less emotionally trying, to be a witness while watching the nightly News on TV, than when reloading your weapon during a battle on the front lines.


I told some friends in Belgium today ...

Today might therefore be a very good day to say a small prayer and a word of thanks to an elderly person in Belgium who is now gone.


Perhaps as breaths and heart beats descended quietly (I hope) towards their last, without even being tested for SARS-CoV-2, and intentionally so in this country, this person you are now praying to, quietly and courageously accepted a fate that was perhaps destined originally for someone else. Someone a few years younger.


As I suggested to my friends, perhaps this solitary goodbye took place alone in a room that had your name written on its door.





I'm probably overstating this. We're talking about only 2,586 old folks. Only 100 more dying in their Rest Camps, than in the hospitals. And if they in fact did not die of COVID-19, well all the better. The hospitals here are already full of the stuff.



The elderly









Our view of the world's COVID-19 challenge this day (in English) >>>>>>>>>>


Same view of the world situation but en français >>>>>>>


Les données pour Belgique, hier >>>>>


Yesterday's report on Belgium, but in French >>>>>>


And finally, yesterday in Belgium, in English >>>>>



<<<<<<< Home




0 Poster un commentaire

Inscrivez-vous au blog

Soyez prévenu par email des prochaines mises à jour

Rejoignez les 13 autres membres