Talk:COVID-19 pandemic in Belgium

Why quote the pages in Flemish here?
The quoted pages on www.info-coronavirus.be are available in Netherlands, French and English. As this pages is intended for native English speakers, the politeness would be to quote pages in that language when they are available. For example, https://www.info-coronavirus.be/en/news is better here than https://www.info-coronavirus.be/nl/news/ even if there is a delay for the English version. — Preceding unsigned comment added by Rpilotte (talk • contribs) 11:31, 10 March 2020 (UTC)
 * Feel free and adapt it in this way. And by the way: the quotes are in Dutch, not in Flemish. And "Netherlands" doesn't not exist as a languange. --Tfa1964 (talk) 23:13, 11 March 2020 (UTC)
 * Only the Dutch and French version of this government site are regularly updated, whereas the English and German part only get sporadic updates, and not all articles are translated (e.g. look at the site, as of time of writing, the English and German part say nothing about the 3 deaths, whereas the Dutch and French version do). Therefore, quoting the site in Dutch and French would be better in my opinion. I get what you mean with this, but knowing how relevant information, especially for this pandemic, comes in fast, it's a necessary evil to keep the page updated to it's fullest extent Kef274 (talk) 07:34, 12 March 2020 (UTC)

WikiProject COVID-19
I've created WikiProject COVID-19 as a temporary or permanent WikiProject and invite editors to use this space for discussing ways to improve coverage of the ongoing 2019–20 coronavirus pandemic. Please bring your ideas to the project/talk page. Stay safe, -- Another Believer ( Talk ) 17:33, 15 March 2020 (UTC)

Wrong graph of deaths?
According to these sources: COVID-19 - EPIDEMIOLOGISCH BULLETIN NL and COVID-19 - BULLETIN EPIDEMIOLOGIQUE FR  the graph 'Confirmed deaths per day in Belgium' is not correct. What is the cause? Where the reported statistics in the past wrong and is this corrected by governement afterwards? Or what is the source of the current graph? --PJ Geest (talk) 09:19, 21 March 2020 (UTC)


 * I agree and I have verified the numbers on the archived copies of the official bulletin in french and in dutch, as well as the press conferences now added daily on youtube, and corrected the counts accordingly. However it feels like we have too many redundant copies of the same numbers, I had to correct 3 tables/charts. --Redturnips (talk) 16:25, 21 March 2020 (UTC)
 * The numbers on the press conference are the additional confirmed deaths each day, but the date of these deaths can be different. In the Sciensano graph, the deaths are according to the effective date of death. Shouldn't we take these dates? --PJ Geest (talk) 20:48, 22 March 2020 (UTC)

Raw data IC patients
Concerning these two sources: COVID-19 - EPIDEMIOLOGISCH BULLETIN NL and COVID-19 - BULLETIN EPIDEMIOLOGIQUE FR.

I would like to have the raw numbers on the first graph on page 3. I would also like to now if the IC numbers are the amount of current IC cases. I.e. if they are corrected for people who got out of IC.

Using this data I want to construct two graphs: - Amount of current patients in IC per day. I believe this is a better reflection of the current hit our healthcare system is taking. - Amount of new patients in IC per day. I believe this is a better reflection of the total amount of people infected. As you would expect a certain percentage of people infected to be hit hard enough to be held in IC.

I believe these two graphs would be better indicators than the amount of confirmed cases, as these numbers highly depend on how much and whom the government is testing. — Preceding unsigned comment added by Jeffreydr (talk • contribs) 12:34, 22 March 2020 (UTC)

Logistic growth model fit for confirmed COVID-19 cases in Belgium for data from 07/03/2020 until 22/03/2020.
I think that it does not make any sense present the forecast over 1 month (using a logistic fit) without also including the uncertainties in the forecast which are likely to be very large. — Preceding unsigned comment added by 217.136.237.223 (talk) 13:16, 23 March 2020 (UTC)

I still find it odd that we can only get the data from 1 day before. Meaning that if you didn't save a bulletin, you lack datapoints. So if anyone has the bulletins saved over the course of a few weeks, I would very much like to get them — Preceding unsigned comment added by Jeffreydr (talk • contribs) 13:03, 25 March 2020 (UTC)

Historical PDF now available as on the French and Dutch page from march 14th onward : https://epidemio.wiv-isp.be/ID/Pages/2019-nCoV_epidemiological_situation.aspx — Preceding unsigned comment added by 195.26.3.225 (talk) 12:25, 26 March 2020 (UTC)

There is also a graph showing a logistic fit on the daily numbers (daily new deaths, daily hospitalized). This doesn't make sense to me. A logistic model can only be used on the totals. Since the logistic model trends towards a certain number, we can't expect to keep having 80 deaths per day due to Covid-19... For the initial phase, logistic growth is just exponential, and you can indeed fit both stats with exponential functions (as the derivative or the integral of an exponential is again an exponential). But once you get into the flattening part of the logistic function, the derivative becomes quite different. Sanderd17 (talk) 10:18, 29 March 2020 (UTC)
 * This chart also doesn't make any sense to me. The last thing we need right now is poorly fitted charts on a wikipedia page. There is a lot of uncertainty when using predictive models for this outbreak, and fitting a curve is definitely not the best predictor. Especially since we've taken measures which alter the basic reproductivity factor. So all in all I'm really struggling to see the benefit of having this chart here. It is not the best predictor by a long shot, it doesn't include a lot of factors, but still gives some misguided information that the outbreak will stop at 9k. --2A02:2788:468:A68:9110:1FFB:C50B:6A96 (talk) —Preceding undated comment added 11:18, 1 April 2020 (UTC)
 * The daily new deaths are not on this graph anymore but I also still find this chart is odd. Jrm7 (talk) 08:32, 2 April 2020 (UTC)

Hospitalized figures wrong.
What is definition ? Total people hospitalized over time or at a specific day ? Figures seems to be mixture of both. Or did you copy the wrong text on the summary part of the bulletin of today ? The line on page 3 is the correct one!!

regards, theo


 * Which part of the article do you think is wrong? --Jrm7 (talk) 00:56, 29 March 2020 (UTC)


 * The title in the wikipage suggests the hospitalized numbers are from people who entered the hospital. But the numbers in the table (and the bulletins) are also corrected for the amount of people who left the hospital. i.e. the value "number of hospitalized people" is the amount of people in the hospital on that day, not the cumulative of the amount of people who ENTERED the hospital over all days. --Jeffreydr (talk) 12:11, 29 March 2020 (UTC)


 * I see, indeed. Well, I think it makes more sense if the total column mentions the cumulative amount of people who where hospitalized, what do you think? Jrm7 (talk) 10:48, 29 March 2020 (UTC)


 * The table is now update and rearranged for more clarity. There is probably also something to fix with the recovered cases as the daily reports only talk about patients who left hospital. We don't know anything about the recovered people who were tested positive but sent home without hospitalization. Jrm7 (talk) 12:32, 29 March 2020 (UTC)
 * As far as I know the recovered cases are only counted coming from an hospital, in the press-conference they often mention "left the hospital"
 * Yes, indeed... Column titles were corrected in that way ('recovery' replaced by 'hospital discharge') and a note was added both in the bottom of the table and in the article. Jrm7 (talk) 23:55, 30 March 2020 (UTC)


 * The graphs with hospitalization data were also updated for more clarity Jrm7 (talk) 22:56, 29 March 2020 (UTC)

A similar problem is the "total" column under ICU. The report contains the number of ICU beds occupied on a given day. Adding them up gives no meaningful figure, as some people occupied a bed for 2 weeks...The day load has to be compared with the number of ICU beds available, to give an idea of the pressure on the health system.


 * There is no "total column under ICU" and ICU cases are in no way added up. The main title of the column is not "ICU" and the data is not about new ICU and/or total of ICU. The main title of the column is "Active" (which refers to the active cases in hospital on each day). The "Active" column is then subdivided with the active ICU on each day and the total of actives cases on each day. It is exactly how data are mentioned in the reports. Jrm7 (talk) 23:47, 30 March 2020 (UTC)

Actual dates vs Report dates
From 31 March Sciensano started to make a set of raw data available, on a daily basis just like the daily reports. Those are the up-to-date data used by Sciensano to study "the dynamics of the epidemic, help to anticipate different scenarios and to elaborate possible measures to curb the spread of the virus" (their words). The shared data are kinda rich and present all the numbers (cases, hospitalisations, deaths, etc.) by actual dates and not report dates, e.g.: the actual date of death and not the date on which the death was reported to Sciensano. Many "numbers" are actually reported with delay as it is explained in the daily reports. Sciensano make the retrospective consolidation and work with the actual dates. They obviously follow the development of the COVID-19 in Belgium based on those actual dates; all the charts in their daily reports are made with the actual dates. In fact, as disconcerting as it can be, the "new" numbers presented in the daily reports do not mean a lot, especially for new cases and new deaths which are mostly reported with delays. Numbers regarding hospitalisations are more reliable bc there is less delay (less retrospective consolidation) but actual dates mostly are just the day before it was reported by Sciensano, as well as all cumulated totals. Given the above, I started to make graphs similar to the ones Sciensano present, considering the actual dates of cases. Now, the question is: Should we keep going with graphs made with dates of reports, and what would be the point? Jrm7 (talk) 16:29, 2 April 2020 (UTC)
 * I would remove the old graphs and only build graphs using the new data, since these will show the actual trends and not trends that are their because of reporting delays etc. Tverhoeven (talk) 08:13, 3 April 2020 (UTC)
 * I would also use the 'actual dates' (not the reporting dates). I now see both 'actual date' and 'effective date' mentioned. Are these the same ? --Tfa1964 (talk) 07:39, 4 April 2020 (UTC)
 * It is supposed to be the same, yes. Effective date is used regarding hospitalisations data. Actually, the xAxis name could be just "Date", I mentioned actual date to highlight the difference with the former graphs using the report date... but they were now deleted. — Preceding unsigned comment added by Jrm7 (talk • contribs) 09:33, 4 April 2020 (UTC)

Open data
To little fanfare, Sciensano released their numbers and a beautiful interactive dashboard. Spreading the word so that we don't have to keep using those awful PDFs.


 * https://epistat.wiv-isp.be/Covid/
 * https://datastudio.google.com/embed/u/0/reporting/c14a5cfc-cab7-4812-848c-0369173148ab/page/tpRKB

Adys (talk) 17:23, 4 April 2020 (UTC)


 * Thank you, this was already linked in the article. Tables and graphs were also updated a few days ago, based on those data. Jrm7 (talk) 19:07, 4 April 2020 (UTC)

Change in death statistics
I heard that Belgium changed the strategy concerning the numbers of covid19-dead a few days ago, sorry, that my English is not good enough to explain it. Before, a lot of dead are counted, even without positive test, different from other countries, but they changed that, the numbers have slowed down the last days, I did not find it mentioned in the article, does anyone have a source for that? --188.97.169.53 (talk) 18:13, 21 April 2020 (UTC)

>Death statistics did not change, be were refined. I added new section around this criticims: [] --Sanderd17 (talk) 08:30, 29 April 2020 (UTC)

Timeline section
The timeline section is not encyclopedic to me. If it's just an enumeration of deaths, it belongs in the statistics part. In my opinion, the timeline and government response can be merged/rewritten. It can still be in a chronological order, where the governement and public responses are a result of the evolution of the number of cases and the mortality.

It could be split in a few phases, each mentioning their general evolution in cases/deaths, combined with what governmental actions were taken as a response, and how the numbers changed. E.g. the following structure could be used:


 * Background: virus in China: Belgian news coverage around it and governmental preparations
 * First case: Handling of the first case on Belgian ground coming from China (did not cause an outbreak)
 * Initial outbreak: Documenting the imported cases from ski holidays and carnival events that did cause an outbreak. Including government responses regarding testing policies or banning trips to Italy.
 * Spread and lockdown: The evolution towards uncontrolable spread and the complete lockdown
 * Peak and exit strategy: When the (first) peak was reached, and how the exit was planned
 * Exit: Whatever comes next...

--Sanderd17 (talk) 07:35, 30 April 2020 (UTC)


 * It is true the Timeline section should be rewritten and mostly strongly summarised. That's the reason of the Template:Recentism (WP:RECENTISM) on top of the section (+ Wikipedia is not a newspaper - WP:NOTNEWS). But it has no sense to merge the Timeline section with the other sections, and to change the whole structure of the article: the exact opposite was done a few weeks ago when the article was a real mess mixing all kind of informations only written in a kinda-chronological order. The actual general structure of the article is now clear and match the pages concerning other countries (like UK or USA). It has no sense and it would bring confusion to split the government response in different parts of the article, especially when the responses and impacts are spread over time like they are: if someone wants to know what the Belgian government did, all this information should be found in a same place/section, and not be forced to read the whole article and to collect parts of information from everywhere in the article. The Timeline section probably tends to disappear in the future as it follows too much a newspaper way of presenting facts, which was ok at the beginning of the crisis but need to be placed in perspective afterwards. In the meantime, I think the section should be seen as a sort of summary of the pandemic. Jrm7 (talk) 09:38, 3 May 2020 (UTC)


 * I just read a few timeline sections. The one for the Netherlands has a number of bullet points like our timeline (with numbers of infected and deaths), but also mentions a lot of governmental actions (advices given, public talks, ...) This would also be welcome in our timeline. And if the virus just spreads as usual, there's no reason to mention how many people were infected in what villages. It's just normal progress so those dates should be skipped in the overview. The French timeline weirdly doesn't read like a timeline, but talks about the different hotspots, about the lock-down, ... I don't think this format is applicable for Belgium. On the UK page, I don't even see a "timeline" section. The government response is just arranged roughly chronological, with little information on the number of cases (or more important, the stage the epidemic is in). I also like this style, as it gives a rough overview of the state a the country was in. But I prefer to have some status of the epidemic mixed in it. The US timeline is a bit more like the Netherlands (though not in bullet points), and it mixes numbers of cases with governmental action.
 * For me, the easiest way to read an intro to a historical topic (which is what this page will become), is when I get in broad lines the action and reaction dynamics. So the epicemic goes into a certain stage, the general public or the government reacts on it, the eidemic goes into another stage, ... I believe that a little country as Belgium is very well suited for this pure chronological action/reaction writing since the viral stages were pretty uniform over the entire country. We don't have to document a number of distinct clusters. --Sanderd17 (talk) 14:52, 4 May 2020 (UTC)


 * The Timeline sections are the remains of how the articles started to be written in the very beginning. It tends to disappear, or to be split, or to be strongly summarised/rephrased, in every page about the pandemic. Especially, as you say, the mentions like new cases in villages and all the small talk we have on the Belgian page indeed are not very useless and not of big interest in a general historical perspective. This being said, the vast majority of the articles – general (COVID-19 pandemic) and by country (e.g. USA, UK, Netherlands, Sweden, Brazil, Russia, China, Singapore, Italy, Spain, Canada, etc.) have all the same common structure by topic, with the dedicated (!) sections: Responses, Impacts, Statistics. The USA and UK pages have well both a Timeline section and this structure by topic (Responses/Impacts/Statistics) + They even have a separate page with their detailed Timeline (Timeline page USA and Timeline page UK). There is no reason to not have this same clear structure and table of contents for the Belgium page, no matter how small we are; the kinds of responses Belgium takes are not less important or "smaller" than other countries. And making a general structure only in a chronological order would be a step back in the evolution process of this page. Furthermore, the action/reaction you're talking about seems a bit hazardous to me because there's not always direct and/or objective and/or systematic links to be made between all the different responses and the moment they were decided or they occurred, except the simple evolution of the situation. Though, I agree the Timeline section has to be fully rewritten, and probably a few words about the context could be added when explaining the responses/measures (as it is already the case sometimes, e.g. "as a second case of coronavirus was confirmed in Belgium, phase 2 of the health risk containment strategy was activated"). Some parts of the Responses sections also need to be rephrased. But the general structure and table of contents with the dedicated sections by topic (Responses/Impacts/Statistics) have to be kept separate (but need to be developed). Jrm7 (talk) —Preceding undated comment added 18:00, 5 May 2020 (UTC)

Problem with deaths in hospital... where do they come from?
Hi, little problem with the total numbers! Hospitalized: 15.000 Discharged: 12.000 Remaining in hospital: 3.000 Right and checked! ... but also Died in hospital: 3.500? Did they die just before entering or just after leaving, but already/still in the hospital? :D (sorry not able to sign, forgot password62.235.80.118 (talk) 14:11, 5 May 2020 (UTC)

Death count comparisons
It is stated that most other countries only count deaths occurring in hospitals. This may have been true at one time, but I'm not sure it is true now (8 May). The UK and France both now count deaths occurring in care homes as well as hospitals. This resulted in a large increase in both countries' death totals, though they are still below Belgium in proportion to population. I believe some other European countries do the same.31.48.173.36 (talk) 18:33, 8 May 2020 (UTC)


 * The claim that "most other countries [...] publish figures based on confirmed hospital deaths only" indeed requires substantiation. Skimming current news articles, the death counts from several European countries do include deaths in care homes, including Germany, France, Italy and Ireland.109.136.43.186 (talk) 10:37, 13 May 2020 (UTC)


 * The text was changed (not by me) to "most other countries, which publish figures based primarily on confirmed hospital deaths" and the "citation needed" note was removed. It is still a claim that requires substantiation and the source linked at the end of the sentence does not support it. It does mention three examples: "In Germany, Italy, Spain and elsewhere, daily death tolls are hospital-only fatalities, which don't include those in care homes and in the community", although other sources such as contradict this at least in part: "France and Germany, for example, have been including deaths in care homes in the headline numbers they produce every day." 109.136.43.186 (talk) 10:32, 27 May 2020 (UTC)

Comparison of excess mortality and reported COVID-19 deaths
"between 5 March and 5 April, Belgium reported 2,373 COVID-19 deaths with the excess mortality being estimated around 3,000. Causing still an underreporting of around 600 deaths. While the neighbouring Netherlands had an estimated excess mortality of 6,200, with 3,197 reported COVID-19 deaths in that same period. Causing an underreporting of around 3000 deaths." There are some problems with this phrasing. Primarily, terming the difference between excess mortality and reported COVID-19 deaths *underreporting* presumes that all excess mortality is due to COVID-19, which is not a reasonable assumption. I suggest to remove the two short sentences that refer to the difference as underreporting.109.136.43.186 (talk) 10:50, 13 May 2020 (UTC)

median age of population versus Corona infects or deaths in ppm?
if we are on correlations, world comparison? Wikistallion (talk) 16:16, 9 May 2020 (UTC)

% positive in the table under "by day" is wrong
I noticed that the % positive tests keeps dropping and dropping, and I noticed that it is not representing what I think it is representing.

Here it calculates new cases / total tests. Shouldn't this be new cases / new tests? I'm not 100% sure and I don't want to edit if I'm not correct, but I'm pretty sure this is calculated wrong. — Preceding unsigned comment added by 94.110.151.31 (talk) 09:35, 15 May 2020 (UTC)


 * I agree with you: it should be new cases / new tests and not new cases / total tests. The whole table needs an update anyway. Jrm7 (talk) 12:07, 18 May 2020 (UTC)

Mortality peak
Mortality peak was two days earlier on 10th of April. Just look at the figures! — Preceding unsigned comment added by 188.101.85.156 (talk) 07:23, 30 May 2020 (UTC)

Death count
What is the overall death count for January 1st to end of May? How does this figure compare to the last years, especially 2018? Did 2020 had more deaths than 2018? — Preceding unsigned comment added by 92.209.196.140 (talk) 08:20, 8 June 2020 (UTC)


 * The article states the death toll is unknown, the national board of statistics are only going by their own estimates. Belgium appears to be a hard-hit nation, although Italy and Spain would have more percentage of population infected yet untested. Belgium unlike Italy and Spain in the winter months has an oceanic climate (closer to the North Sea) instead of a cold, dry one and less people are outdoors in those months compared to warmer countries. The largest concentration of cases are in the tri-national (Belgian/Dutch/German) Gangelt region, one of Europe's first and heaviest clusters of COVID-19 positive cases. 2605:E000:100D:C571:A8BB:CE5:5FFF:7B6A (talk) 18:19, 22 June 2020 (UTC)

source ?
@Ibisbird,

why don't you name your sources for your sentences ?

I have changed it and added two sources. --Neun-x (talk) 19:38, 24 September 2020 (UTC)

Request for comment - cases change metrics
Please take part in discussion here: Project COVID-19, Medical cases charts - change type — Preceding unsigned comment added by Kohraa Mondel (talk • contribs) 23:08, 5 December 2020 (UTC)

Belgium: Highest death rate AND second highest rate of infection
Fresh insights on the situation in this country; see the final sections especially. AND NOTE THIS Belgium has the second-highest rate of infection in Europe behind the Czech Republic. That indicates that the high death rate is understandable. https://dispatcheseurope.com/expat-essentials-dispatches-complete-guide-to-pandemic-rules-restrictions-across-europe/  AND see https://www.cnn.com/interactive/2020/health/coronavirus-maps-and-cases/

Belgium has the dismal distinction of having the highest rate of deaths per million of population – 1,565 as of 15 December – in the world and more than 18,000 deaths since the pandemic began.https://dispatcheseurope.com/expat-essentials-dispatches-complete-guide-to-pandemic-rules-restrictions-across-europe/  AND also see https://www.cnn.com/interactive/2020/health/coronavirus-maps-and-cases/

Why such a high death rate? Although it is hard to fully explain the different death rates among countries, certain explanations have been put forward more than others. Indeed, as COVID-19 casualties are higher in older patients, countries could face different death tolls due to the different age structures of their population. For instance, 70 percent of deaths in the Netherlands occurred in people aged over 70, and 51 percent of deaths in Belgium occurred in nursing homes. '''However, in 2020, the Belgian population was not particularly old, especially when compared to countries such as Germany or Japan. Therefore, experts are considering further reasons to fathom such a high death rate.'''

Other explanations As of December, over one million people have been tested positive for the coronavirus in the Benelux, from which 609 thousand in Belgium and 628 thousand in the Netherlands. These numbers solely reflect confirmed and diagnosed cases and not the entire number of sick or infected people. Indeed, numbers depend on the logistic capacity for testing, which can vary from one country to another. The accurate number of Sarscov2 infected people is, therefore, unknown. Meanwhile, the number of deaths due to the virus relies on this unknown number. Moreover, some deaths are not accounted for in the death tolls because not all deaths have been officially linked to the coronavirus. For instance, in France or Italy, COVID-19 death tolls solely include tested casualties while untested deaths are not systematically considered. In Belgium, however, suspected cases are accounted for in the death toll, regardless of the person having been tested or not. HOW GREAT IS THE OVER-REPORTING????https://www.statista.com/statistics/1107558/coronavirus-death-casualties-in-netherlands-belgium-luxemburg-benelux/ Peter K Burian (talk) 18:04, 17 December 2020 (UTC)

Infections at the Doel Nuclear Power Station
Hello, is that relevant for the article? I assume there are much more 'critical' installations that got a high number of infections at a certain moment in time ... SpamHunters (talk) 14:23, 26 February 2021 (UTC)

Semi-protected edit request on 7 January 2022
Is it possible to have frequent updates ???? It is not updated since December 8, contrary to all big countries where it is updated every day !!!!!!!!!!!!!! Thanks. 2A02:A03F:C07D:C200:F8CF:1C7E:42BC:B60C (talk) 11:19, 7 January 2022 (UTC)
 * Red question icon with gradient background.svg Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. ScottishFinnishRadish (talk) 12:44, 7 January 2022 (UTC)